F. J. Díez, M. Luque, C. L. König and I. Bermejo. Decision analysis networks. Technical Report CISIAD-14-01, UNED, Madrid, 2014.
11 pages. PDF (564 KB), BibTeX entry.
This paper presents decision analysis networks (DANs) as a new type of probabilistic graphical model. Like influence diagrams (IDs), DANs are much more compact and easier to build than decision trees, and are able to represent conditional independencies. Both IDs and DANs can represent symmetric problems, but DANs can also represent problems involving restrictions between the values of the variables (structural asymmetry) and partial orderings of the decisions (order asymmetry). Therefore, DANs can easily model and solve many real-world problems that IDs cannot. We argue that DANs compare favorably with other formalisms proposed for modeling asymmetric decision problems. Finally, we offer OpenMarkov as an open source software tool for representing and evaluating DANs.